Blog / model configuration
Hermes Agent with Nous Portal: One Login, Many Models
Nous Portal is the subscription gateway built by Nous Research, and it is the shortest path to a working Hermes Agent because one login covers both the models and the common tools. Instead of collecting an API key for every lab, you authorise once, pick a model, and the assistant can search, generate images, and transcribe without extra credentials.
- 01
Confirm the subscription is active
Make sure the Portal subscription is live before you start, so the first request is a configuration test rather than a billing diagnosis.
- 02
Complete the Portal login
Run the interactive setup and finish the browser-based authorisation. The flow registers the credential for you, so there is no key to copy or store by hand.
- 03
Pick the default model
Choose one model as the day-to-day default. Keep it explicit so a change on the provider side cannot move the assistant onto something you did not choose.
- 04
Turn on the tool gateway
Enable the bundled tool access so web search, image generation, and browser automation work without separate provider accounts.
- 05
Verify a request and a model switch
Send one normal prompt, then switch to a different model mid-session and send another. Both working means the login and the catalog are correctly wired.
Four-step path
How do you set up Hermes Agent with Nous Portal?
Four checkpoints: confirm the subscription, complete the login, choose a model, then verify that tools and model switching work.
- Step 01
Confirm the subscription is active
Make sure the Portal subscription is live before you start, so the first request is a configuration test rather than a billing diagnosis.
- Step 02
Complete the Portal login
Run the interactive setup and finish the browser-based authorisation. The flow registers the credential for you, so there is no key to copy or store by hand.
- Step 03
Pick the default model
Choose one model as the day-to-day default. Keep it explicit so a change on the provider side cannot move the assistant onto something you did not choose.
- Step 04
Turn on the tool gateway
Enable the bundled tool access so web search, image generation, and browser automation work without separate provider accounts.
- Step 05
Verify a request and a model switch
Send one normal prompt, then switch to a different model mid-session and send another. Both working means the login and the catalog are correctly wired.
What is Nous Portal?
Nous Portal is a unified subscription gateway for Hermes Agent that replaces per-vendor accounts, keys, and billing with a single authorised login.
The gateway curates a broad catalog of agentic models and routes each one to the backend best suited to it. Routing decisions can change over time, which is usually an improvement rather than something to pin down.
What does the subscription actually include?
It covers a large frontier model catalog across the major labs plus hosted tool access, so one subscription replaces both an inference account and a handful of tool API keys.
- Frontier models from several labs under one entitlement.
- Hosted tools such as web search, image generation, and browser automation.
- One bill instead of a separate balance per provider.
- Mid-session model switching without new credentials.
How is it different from bringing your own keys?
Bring-your-own keys give you direct control of each vendor relationship; the Portal gives you one relationship and less setup. The right choice depends on whether you value control or simplicity more.
Teams that already have negotiated provider contracts should keep them. Teams that just want the assistant working today are better served by a single login and a single invoice.
How do you switch models later?
Switch with the model command or the model picker in the web dashboard. Because the credential is account-wide, changing models needs no new secret and no redeployment.
- Use a strong reasoning model for multi-step and tool-heavy tasks.
- Use a faster, cheaper model for summaries and classifications.
- Re-test tool calling after any model change.
- Keep the default stable for shared deployments so answers stay consistent.
What should you verify after connecting?
Check that a plain prompt answers, that a tool-backed request works, and that usage appears in the Portal account, because those three together confirm authentication, entitlements, and billing.
- A plain prompt proves the login and model access.
- A tool-backed request proves the bundled tool access is enabled.
- Usage appearing in the account proves billing is attached to your subscription.
- A second model answering proves the catalog is not limited to one family.
Which primary sources support this page?
Product behavior is checked against the current deployment flow. These external links provide the upstream project or channel documentation used for setup details.
Direct answers
Frequently asked questions
Do I still need provider API keys with Nous Portal?
No for the models and tools the subscription covers. Keep a direct key only if you rely on a provider-specific feature outside the catalog.
Can I mix Portal models with my own keys?
Yes. Many teams keep the Portal as the default and a direct key as a fallback, so a subscription or quota problem does not stop the assistant.
Does switching models mid-session lose context?
The conversation continues, but a different model may interpret the existing context differently. Switch at a natural break rather than mid-task.
Is the subscription cheaper than paying each lab directly?
It depends on your usage mix. Compare against your current per-vendor spend, and remember that consolidating also removes setup and admin time.
Key takeaways
- Nous Portal is the subscription gateway built by Nous Research, and it is the shortest path to a working Hermes Agent because one login covers both the models and the common tools. Instead of collecting an API key for every lab, you authorise once, pick a model, and the assistant can search, generate images, and transcribe without extra credentials.
- Start with one model, one channel, and one controlled test conversation.
- Use managed hosting when deployment speed matters more than operating the server yourself.
Continue the setup
Related Hermes Agent guides
Blog guide
How to Choose a Model for Hermes Agent
Choose the right Hermes Agent model for support, briefings, operations, and channel workflows across GPT, Claude, Gemini, DeepSeek, Kimi, and GLM.
Read nextBlog guide
Hermes Agent Provider Routing Across Multiple Models
Route Hermes Agent across multiple model providers by task, cost, and latency. Covers rule design, verification steps, and the failure modes to avoid.
Read nextBlog guide
Hermes Agent Pricing and Cost: What You Actually Pay
Compare Hermes Agent Starter and Pro pricing with model, VPS, maintenance, and failure costs before choosing managed hosting or self-hosting.
Read nextBlog guide
Hermes Agent OpenRouter Setup: One Credential, Many Models
Connect Hermes Agent to OpenRouter with an API key or browser login, then control routing, fallbacks, and spend across many model labs from one place.
Read next